English

An Entropy Search Portfolio for Bayesian Optimization

Machine Learning 2015-03-06 v4 Machine Learning

Abstract

Bayesian optimization is a sample-efficient method for black-box global optimization. How- ever, the performance of a Bayesian optimization method very much depends on its exploration strategy, i.e. the choice of acquisition function, and it is not clear a priori which choice will result in superior performance. While portfolio methods provide an effective, principled way of combining a collection of acquisition functions, they are often based on measures of past performance which can be misleading. To address this issue, we introduce the Entropy Search Portfolio (ESP): a novel approach to portfolio construction which is motivated by information theoretic considerations. We show that ESP outperforms existing portfolio methods on several real and synthetic problems, including geostatistical datasets and simulated control tasks. We not only show that ESP is able to offer performance as good as the best, but unknown, acquisition function, but surprisingly it often gives better performance. Finally, over a wide range of conditions we find that ESP is robust to the inclusion of poor acquisition functions.

Keywords

Cite

@article{arxiv.1406.4625,
  title  = {An Entropy Search Portfolio for Bayesian Optimization},
  author = {Bobak Shahriari and Ziyu Wang and Matthew W. Hoffman and Alexandre Bouchard-Côté and Nando de Freitas},
  journal= {arXiv preprint arXiv:1406.4625},
  year   = {2015}
}

Comments

10 pages, 5 figures

R2 v1 2026-06-22T04:41:08.327Z